You are on page 1of 32

This article challenges the widespread notion that civic nationalism is dominant in Western

Europe and North America, whereas ethnic nationalism is dominant in Central and Eastern
Europe. After laying out the civic-West/ethnic-East argument. it refines the civic/ethnic dichotomy and deduces the state policies that flow from ethnic, cultural, and civic conceptions of
national identity. It then employs survey data from 15 countries to measure mass conceptions of
national identity by analyzing attitudes on criteria for national membership and state policy
toward assimilation and immigration. The article finds that the civic-West/ethnic-East stereotype, when true, is only weakly true, and according to several measures is false. Finally, several
explanations for strong cultural national identities in the West and strong civic national identities
in Eastern Europe are given.

CHALLENGING THE CIVIC/ETHNIC


AND WEST/EAST DICHOTOMIES IN
THE STUDY OF NATIONALISM

Southern Illinois

STEPHEN SHULMAN
University at Carbondale

he distinction between civic and ethnic nations is one of the most


widely employed conceptual building blocks in the study of ethnic relations and nationalism. A closely related distinction is that between Western
and Eastern nations, according lo which Western Europe and the United
Statesdeveloped primarily ascivic nations and Germany and Eastern Europe
primarily as ethnic nations. Unfortunately, few analyses challenge these
dichotomies theoretically or empirically. This article seeks to assess the
accuracy of the conventional wisdom that fundamentally accepts the civicWest/ethnic-East pattern of national development. Using survey data from 15
countries, it arguesthat the standard view greatly exaggeratesthe current differences in national identity between the West and East. Western civic
nations are more ethnic than is usually recognized, and Eastern ethnic nations
AUTHORS NOTE: I would lik to thank Torus

Kuzio,fur

his cwntnents on an em&v

this urticle.

COMPARATIVEPOLITICAL STUDIES,Vol. 3.5No. 5, June2002 554-585


0 2002 Sage Publications

554

version of

Shulman / CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

555

are more civic. Furthermore, on some measurements, countries in Central


and Eastern Europe are more civic and less ethnic than Western countries.

THE CIVIC/ETHNIC AND WEST/EAST


CONCEPTUAL DICHOTOMIES
For much of the 20th century, scholars of nations and nationalism have
commonly argued that (a) there are many different traits that can provide the
foundation for national unity and identity, and (b) nations differ in the mix of
the traits that form the basis of their unity and identity. A simple classificatory
scheme has arisen that distinguishes nations as civic, political, or territorial
on one hand, versus ethnic or cultural on the other. Parallel with this conceptual distinction is a geographic one in which civic components of nationhood
are dominant in Western Europe and the United States,whereas ethnic components are dominant in Central and Eastern Europe.
In his 1907 work Cosmopolitanism and the Natiorlal State, German historian Friedrich Meincke (1970) becameone of the first scholars to asserta fundamental difference between political and cultural nations. However, it was
Czech emigre Hans Kohn who developed and popularized the dichotomous
framework in his 1944 book The Idea of Nationalism and later works (Kohn,
1944, 1946, 1949). Kohn argued that in the West, particularly England,
France, the Netherlands, Switzerland, and the United States,nationalism was
primarily political. There, ideas of the nation and nationalism arose within
preexisting state structures that encompassedpopulations with a relatively
high degreeof cultural homogeneity, or developed simultaneously with those
structures. Inspired by Enlightenment ideas of liberty and equality, Western
nationalism struggled against dynastic rule and equated citizenship with
membership in the nation. Members of the nation were unified by their equal
political statusand their will as individuals to be part of the nation. Thus in the
Western model, the state temporally precedes (or coincides with) the development of the nation. In the socially and politically more backward areasof
Central and Eastern Europe and Asia, however, nationalism arose in polities
that very poorly coincided with cultural or ethnic boundaries (e.g., Russian,
Austro-Hungarian, and Ottoman empires). In these regions, Kohn (1944)
argues, nationalism struggled to redraw the political boundaries in conformity with ethnographic demands (p. 329). Thus in the Eastern model the
nation precedes, and seeks to create, the state. Nations in the East consolidated around the common heritage of a people and the irrational idea of the
volk (people), instead of around the notion of citizenship.

556

COMPARATIVE POLITICAL STUDIES / June 2002

Many contemporary scholars employ the analytical distinction between


civic/political and ethnic/cultural nations originated by Meincke and Kohn,
although they differ in their precise characterization of these concepts. Alter
(1994) contrasts cultural nations, basedon common heritage, language, distinct area of settlement, religion, customs, and history, with political nations,
composed of politically aware citizens equal before the law (p. 9). In Smiths
(1991) Western or civic model, national unity arises from a historic territory,
laws and institutions, the legal-political equality of members that expresses
itself in a set of rights and duties, and a common civic culture and ideology.
Smiths non-western ethnic nation is basedon descent or presumed descent,
and thus is seen by its members as a fictive super-family. In addition, vernacular culture, especially language and customs, is a key clement of the ethnic nation (p, 12).
Ignatieff (1993) characterizes the civic nation as a community of equal,
rights-bearing citizens, united in patriotic attachment to a shared set of political practices and values (pp. 6-7,9). In contrast to this notion, which originated in Great Britain, France, and the United States, Ignatieff explains that
for the German Romantics of the 19th century, national unity sprung not from
the cold contrivance of sharedrights but [from] the peoples preexisting ethnic characteristics: their language, religion, customs, and traditions (pp. 6-7).
In addition, he assertsthat ethnic nationalism tells people to only trust those
of your own blood (p. 9).
Regardless of the differences in definition, the assertion that Western
Europe and the United Statesdeveloped primarily as civic nations and Eastem Europe primarily as ethnic nations is common. But it is not just a historical argument; many scholars seethe continuation of historical patterns in current national identities in the West and East. Particularly with the collapse of
communism in the Soviet Union and Eastern Europe, the conceptualization
has become a standard point of departure for many studies of ethnic relations
and nationalism in the post-communist New Europe. According to
Brubaker (1996), the 20 or so new post-communist statesof Eastern Europe
and the former Soviet Union were conceived and justified, in the nationalist
movements preceding their independent statehood as well as after statehood
was achieved, as the state of and for a particular ethnonational group (p. 65,
note 13). Brubaker thinks that civic notions of nationhood have little chance
of prevailing in the new statesof Eurasia, given the pervasively institutionalized understandings of nationality asfundamentally ethnoculturdlrather than
political, as sharply distinct from citizenship, and as grounding claims to
ownership of polities (p. 105).
Another expert on Eastern Europe, George Schopflin (I 996), also stresses
the strongly ethnic character of nationhood and state legitimation in the

Shulman / CIVIUETHNlC

AND WEST/EAST DICHOTOMIES

557

region (p. 153). Following Kohn, Schopflin argues that in the West citizenship arose before or concomitant with nationalism, whereas in Eastern
Europe elites mobilized nations around ethnicity in an attempt to carve states
from the empires that subjugated them. But in addition to this relatively distant historical source of current ethnic nationalism in Eastern Europe,
Schopflin underscores the role that communist rule played.
Schopflin links communism and ethnic nationalism in the region in several ways. First, because communism was associated with foreign oppression, opposition to communist rule became a national project. In these circumstances, Schopflin (1996) writes, it was very difficult for any civic
dimension of nationhood to emerge (p. 153). Second,communism destroyed
civil society in the region, turning communities there into civic deserts
characterized by mistrust and atomization, Consequently, ethnonational
identities were the only ones in the public sphere that could become salient
(p. 153). Finally, communism pushed out other competing ideas and values,
making it much easier for an undiluted nationalism referring solely to ethnicity to survive more or less intact (Schopflin, 1995, p. 53). As a result,
many national disputes from the pre-communist era were swept under the
carpet, only to reemerge with the end of communism.
A final argument for the alleged predominance of ethnic nationalism in
Eastern Europe concerns the newness of the statesand regimes that are constructing from scratch democratic political and legal institutions. Echoing
Gellner (1992), Snyder (1993) contends that ethnic nationalism fills an institutional vacuum. Thus ethnic nationalism is the default option: It predominates when institutions collapse, when existing institutions are not fulfilling
peoples basic needs, and when satisfactory alternative structures are not
readily available (p. 86). In the context of wholesale institutional transformation in Eastern Europe, with its attendant economic malaise, widespread
corruption, and high crime rates, one would expect ethnically based conceptions of nationhood to be very strong in the region according to this argument.

CRITIQUE OF THE CIVIC/ETHNIC


AND WEST/EAST DICHOTOMIES
The civic/ethnic and West/Eastdichotomies have been criticized, although
not to the extent one would expect given their simplicity and ubiquity. The
criticisms that exist can be categorized as normative, conceptual, and
empirical.
Several scholars criticize the dichotomies for their normative, ethnocentric bias. For example, Yack (1999) writes that the civic/ethnic dichot-

558

COMPARATIVE POLITICAL STUDIES /June 2002

omy parallels a series of other contrasts that should set off alarm bells: not
only Western/Eastern, but rational/emotive, voluntary/inherited, good/bad,
ours/theirs! (p. 105). And McCrone (1998) comments that the civic/ethnic
distinction does lend itself to ethnocentric caricature-why cant they be
more like us? (p. 9).
Conceptually, some scholars have attacked the logic of the civic/ethnic
distinction. Yack challenges the notion that acivic identity must be freely and
rationally chosen, whereas an ethnic identity is inherited and emotionally
based.Civic identities can be inherited too, Yack (1999) argues (p. 109). Furthermore, both Nielsen and Kymlicka note that it is a mistake to equate ethnic
with cultural nationalism, becausethey differ according to their openness to
outsiders. Each scholar points to Quebec and Flanders as nationalisms that
have been labeled ethnic but in actuality are cultural (Kymlicka, 1999, p. 133;
Nielsen, 1999, p. 126). Nieguth (1999) similarly calls for unpacking the
dichotomy, seeing ancestry, race. culture, and territory asanalytically distinct
basesfor national membership.
On the empirical side, most scholars who employ the civic/ethnic and
West/East dichotomies are quick to say that most states and nations conlain
both ethnic and civic components. Smith (1991), for example, writes that
every nationalism contains civic and ethnic elements in varying degreesand
different forms. Sometimes civic and territorial elements predominate; at
other times it is the ethnic and vernacular components that are emphasized
(p. 13). Still, Smith contends that the civic/ethnic distinction remains valid
and useful (p. 81). Calhoun (1997) writes that although France and Germany are commonly seen as typical civic and ethnic nations, respectively,
France and Germany, and all of Western and Eastern Europe, have been
shapedby the international discourse of nationalism-including both ethnic
claims and civil prqjects of popular political participation (p. 89).
Despite such empirical criticisms, there have been no works that have
attempted to systematically measure and compare conceptualizations of
nationhood in Western Europe/North America and Central/Eastern Europe.
To empirically evaluate the civic-West/ethnic-East characterization, it is necessary to refine the civic/ethnic dichotomy and settle on a strategy for measuring national identity.
The main problem with the civic/ethnic dichotomy is that it collapses too
much in the ethnic category. As Kymlicka, Nielsen, and Nieguth correctly
point out, ethnic and cultural components of identity should be distinguished.
Superior to the current dichotomy is a schemewith three variants of what can
be called the content of national identity-factors that people in a nation
believe are, or should be, the most important in uniting and distinguishing
them from others and that become the basis for defining membership in the

Shulman / CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

559

Table 1
Alternative

Contents

of National

Identity

Cootent of National Identity

Key Components

Civic

territory
citizenship
will and consent
political ideology
political institutions and rights
religion
language
traditions
ancestry
race

Cultural

Ethnic

The main variants for the content of national identity are civic, cdturd, and erhnic (see Table 1). The existing literature on civic nations produces five main components of the civic identity, according to which national
unity and membership in the nation derive from attachment to a common territory, citizenship, belief in the samepolitical principles or ideology, respect
for political institutions and enjoyment of equal political rights, and will to be
a part of the nation. Cultural identity is based on nonpolitical cultural traits.
The key components here are language, religion, and traditions. Finally, for
ethnic national identity, shared ancestry and race are the dominant criteria by
which membership in the nation is defined.
To someextent, the three variants of national identity differ in their level of
inclusiveness. It is very difficult for outsiders to meet the ethnic criteria
becauseone cannot choose or change ones genesor ancestors.But it is possible to adopt cultural traits and thereby enter the nation. Some of the civic criteria are indeed relatively easy to meet, such as will and consent, but citizenship dependson the choice not of the prospective member of the nation but of
the state. In addition, as will be discussed in more detail below, attachment to
territory can be an exclusive characteristic that is hard or impossible for outsiders to acquire.
Beside refining the civic/ethnic framework, evaluating the civic-West/
ethnic-East argument requires measuring and distinguishing types of nationhood. Two main options exist: examining the policies of statesor the attitudes
of members of the nation. Scholars often interpret a states policies as supportive of the construction of ethnic versus civic nations, and to some extent
belief in the civic-West/ethnic-East pattern is basedon analysis of these policies (see, for example, Brubaker, 1992).
nation.

560

COMPARATIVE POLITICAL STUDIES /June 2002

The problem with inferring national identity from state policies concerning such issues as citizenship, cultural assimilation, treatment of minorities,
and immigration is that to the extent that such policies reflect conceptions of
identity, they embody the views of a very narrow segmentof the states population: the political elite in power at the time the policies were enacted. With a
change in ruling parties or coalitions, policies can change rapidly. A superior
analysis would be one investigating the national identity of either the entire
spectrum of political elites in a country or its mass population. This study
chooses the latter.
Another problem is that state policies may reflect concerns other than
national identity. This is especially the case with immigration policy, which
may also be influenced by concerns over the economic and social impact of
immigration. Nevertheless, massattitudes toward statepolicies can be useful
as an indirect measureof the content of national identity that people support
in acountry. Thus, in addition to directly analyzing massattitudes on national
membership criteria, this article explores masspolicy preferences on cultural
assimilation and immigration.
To do so, however, we must deduce the policies that follow from civic, cultural, and ethnic ideas of nationhood. Again, the lack of inquiry into this matter is puzzling. Extant discussions of the relationship between national identity and state policies are mired in confusion. The primary problem is that
many scholars conflate civic nationhood with cultural assimilation policies.
For example, Brubaker (1992) argues that a policy of cultural assimilation
presupposes a political conception of membership in the nation (p. 8). In a
similar vein, Kymlicka (1999) describes how ethnic conflicts often result
from the attempt by civic nationalists to forcibly incorporate and assimilate
ethnic minorities into the dominant culture (p. 134).
The weaknessof such claims is that no theoretical or logical link is made
between civic nationalism and cultural assimilation of minorities. These
scholars simply label France and other Western nations civic, recognize that
they engaged in assimilation, and therefore conclude that civic nationhood
demands or leads to assimilation. But a truly civic conception of the nation
entails no need for cultural unity. People in a purely civic nation are united by
such traits as common citizenship, respect for law and state institutions,
belief in a set of political principles, and so forth. Similarity in language, religion, and other cultural markers is not necessaryfor the development of such
traits.
Theoretically, the idea of civic nationhood leads to one of two policies in
the cultural sphere (seeTable 2). The first option is a laissez-faire approach in
which the state is as culturally neutral as possible and promotes individual,
not collective, rights. A second option is a policy of multiculturalism. Here

Shulman / CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

561

Table 2
National

Identity and Key Policy Issues

Content of National Identity


Civic

Cultural

Ethnic

Cultural Policy
Promote no ethnic cultures
or promote minority
ethnic cultures
Do not encourage assimilation

Immigration Policy
Open immigration

Entry for all immigrants

Promote dominant ethnic


groups culture
Encourage assimilation

Conditional immigration

Promote dominant ethnic


groups culture
Do not encourage assimilation

Restrictive immigration

Preference for cultumlly


similar immigrants

Preference for ethnically


similar immigrants

the staterecognizes collective rights and promotes the maintenance or development of minority cultures in an attempt to buy ethnic minority attachment
to the state and its institutions and territory. Such an attempt is predicated on
the (civic) assumption that will and consent are central aspects of national
identity. In contrast, nation building under a cultural concept of nationhood
requires that the state pursue cultural assimilation of minorities, becausecultural unity is the foundation for a strong nation-state in this formulation. Here
the state will actively promote the majority ethnic groups history and culture
in education and language policies as the core around which to build a
national culture. Finally, an ethnic conception of the nation also logically
leads to the promotion of the dominant ethnic groups culture, because the
state is conceived here as the stateof and for a particular group, and a groups
common ethnic identity is expressed through its culture. But assimilation is
not encouraged because even if other ethnic groups acquire the majority
groups culture, they can never be part of the latter group, and the ethnic
nation would not be strengthened or expanded in any meaningful way by
such assimilation.
In the realm of immigration, a civic nation should be relatively open to foreigners of any background. The civic nation has a large capacity to absorb
new membersbecauseit makesfew demands on the nonpolitical beliefs and
characteristics of its population. For those immigrants who live in the country
long enough, or their offspring, it is relatively easy to meet the criteria for
national membership. A purely cultural nation should also admit immigrants,

S62

COMPARATIVE POLITICAL STUDIES /June 2002

but only conditionally. Adopting a new culture is not easy and is sometimes
resisted. Therefore immigration should proceed only as long as immigrants
and their offspring are willing and able to assimilate into the dominant culture. Furthermore, a cultural nation will have a bias in accepting immigrants
from culturally similar countries, precisely because this will facilitate their
assimilation into the national culture. Finally, ethnic nationhood calls for a
highly restrictive immigration policy. Immigrants cannot be true members of
the nation because they lack the proper ancestry. Therefore any ethnic
nation that permits large numbers of immigrants decreasesthe numerical
dominance of the majority ethnic group and risks fueling discord and disharmony. The exception here is immigrants who are of the sameethnic stock as
the dominant group in the ethnic nation. The state should permit easy entry to
such immigrants.

ASSESSING THE CIVIC-WEST/


ETHNIC-EAST STEREOTYPE
To evaluate the accuracy of the image of a civic West and an ethnic East,
this article uses a large multicountry public opinion dataset. Under the auspices of the International Social Survey Program (ISSP), in 1995-1996, 23
countries implemented a survey with questions related to national identity.
Fifteen countries were selected for analysis here. Representing the West are
the United States, Canada, Great Britain, Spain, the Netherlands, Norway,
and Sweden. In Central Europe, Germany was selected and divided into its
Western and Eastern components. Analyzing the two parts of Germany separately is necessary, given their disparate postwar histories, and permits a
rough assessmentof the effect of communism on national identity. Finally,
Poland, Hungary, the Czech Republic, Slovakia, Bulgaria, Slovenia, and Latvia were chosen to represent Eastern Europe. Majority and minority groups
in all of these statesare likely to differ in their conceptions of national identity. Consequently, it is necessaryto focus on one or the other for the purposes
of comparative analysis, so that any differences in identity that are found are
not the result of the proportion of minorities in the various countries. This
study will investigate the views of the majority ethnic or ethnocultural
groups.
1. The other European countries participating in the International Social Survey Program
study but not included in this analysis were Ireland, Italy, Austria, and Russia.
2. Identifying members of the majority ethnic or etbnocultural group in each country was
generally not problematic, because most surveys explicitly asked about ethnicity. The excep-

Shulman / CIVUETHNIC

AND WEST/EAST DICHOTOMIES

563

Current conceptions of the civic-West/ethnic-East stereotype lead to the


prediction that civic national identities should be strongest in the West and
weakestin EasternEurope, with Central Europe-the two halves of Germanyfalling somewhere in between. Similarly, ethnic and cultural national identities should be strongest in Eastern Europe and weakest in the West, with Central Europe again in the middle. Germany should, on the whole, be less
ethnically/culturally nationalistic and more civic than Eastern Europe not
only because its Western part did not experience communism, which is
alleged to increase ethnic nationalism. but also due to the greater political
institutional strength and economic prosperity of Germany (its Eastern part
included) than Eastern Europe. In addition, Western and Eastern European
national identities should notjust differ, but differ greatly given the disparate
historical contexts (both recent and distant) that are alleged to influence
national development. Finally, arguments on the influence of communism on
national identity lead to the prediction that Western Germany should be substantially more civic and less ethnic/cultural than Eastern Germany.
DIRECT MEASURES OF NATIONAL

IDENTITY3

The ISSP questionnaire contains a battery of questions explicitly designed


to measurerespondents conceptualization of national identity and membership. Unfortunately, these questions tap into only the civic and cultural bases
of national identity and exclude purely ethnically grounded notions of
national identity. Thus this article cannot offer a full test of the civicWest/ethnic-East dichotomy.
Respondents in each country were asked,Some people say the following
things are important for being [e.g., truly British, Spanish, Hungarian, etc.].4
tions were the United States, Britain, Spain, and Canada. In the United States, Whites were identified m those who said their ancestors came from any European country, Canada, Australia, or
New Zealand. In Britain, the majority group could be identified only by first selecting those who
declared themselves to be English, Scottish, Welsh. or British. From this group were selected
those living in England. Finally, Spaniards and Canadians who speak Spanish and English,
respectively, at home were coded as the majority group.
3. For all calculations in this article, the survey data were weighted using the weighting variable in the dataset.
4. Two exceptions to the phraseology being truly X or being a real X must be noted. In
Slovakia, respondents were asked about the importance of various factors for being a real citizenof theSlovakianRepublic.In Latvia, thephrasearealinhabitantofLatviawas
used. Such
wording does bias the question in favor of civic national identification for these two cases. However, this is offset by a similar bias in three of the seven Western countries. In Britain. Canada,
and the United States, the survey question asks about membership in a group delineated by state
territorial boundaries (British, Canadian. American). In contrast, in tive of the seven Eastern

564

COMPARATIVE POLITICAL STUDIES I June 2002

Others say they are not important. How important do you think each of the
following is . . . ? Respondents were asked to rank the following traits for
national membership as 1 (veq imyortunt), 2 cfairly important), 3 (not very
important), or 4 (not important ut all):
l
l
l
l
l

l
l

To havebeenborn in [respondentscountry].
To havecitizenship in [respondentscountry].
To havelived in [respondentscountry] for mostof oneslife.
To be ableto speak[the dominantlanguagein respondentscountryJ5
To be a [believerin the dominantreligion/denominationof respondentscountry (e.g.,Protestant,Christian, etc.)].
To respectpolitical institutions andlaws of [respondentscountry].
To feel [British, Spanish,Hungarian,etc.].

Of the seven items, those on religion and language evaluate support for cultural national identity, whereas the rest tap into support for civic identity.
Four of the five major components of civic identity can be measured with
these latter items-will and consent, citizenship, political institutions, and
territory. Unfortunately the surveys did not ask a question designed to evaluate the role of political ideology in creating national identity. A question from
a separatebattery of questions permits measurementof the third major component of cultural identity-traditions. Starting with the civic items, we compare the average scoresin each country and in each region (the West, Eastern
Europe, and Central Europe) to evaluate the accuracy of the civic-West/
ethnic-East argument.
The question on citizenship asksabout the importance of state-sanctioned
membership in the civic nation. Those respondents supporting a basically
cultural or ethnic notion of national identity should find such official designation of little importance to national membership. The data show that for all
countries, most ethnic-majority respondents find citizenship an important
component of national identity, as the average scores vary between 1.37 and
1.87 (seeTable 3).6Note that a countrys score must exceed 2.5 in order for its
respondents to find, overall, that the given item is unimportant for national
membership. Within the relatively narrow range of variation among counEuropean cases and Germany, the survey asks about membership in a group that can refer to
either inhabitants of the state or its dominant ethnic group (Pole, Czech. Hungarian, Bulgarian,
Slovenian, German). Thus the net effect of question wording and the distribution of titular ethnic
groups is to slightly bias the survey results in favor of the civic-West/ethnic-East dichotomy.
5. The Canadian survey asked about ability to speak either English or French.
6. The data in all tables exclude the responses Dont Know and Refused to Answer.

Shulman / CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

565

Table 3
Ethnic Majority

Wews on Importance

of Citizenship for National

c0u11try

United States
Canada
Norway
Bulgaria
Great Britain
Slovakia
Sweden
Czech Republic
Slovenia
East Germany
Poland
Latvia
West Germany
Huwary
Spain
The Netherlands
west
Central Europe
Eastern Europe
Note: 1 = very important;

4 = not important

Membership
N

Mean

831
1073
1434
891
822
1207
1344
870
940
574
1506
553
1135
960
968
1990
8462
1709
6927

1.37
1.51
1.53
IS8
1.58
1.62
1.65
1.66
1.70
1.71
1.72
1.75
1.80
1.84
1.84
1.87
1.65
1.77
1.69

at all.

tries for this item, the West (mean = 1.65) places slightly more importance on
citizenship than Eastern Europe (1.69). This supports the relationship
hypothesized by current theories, but the difference between the regions,
although statistically significant, is very small (a later table investigates in
more detail the statistical significance of regional differences). Looking more
closely at the rankings, we find several surprises. Whereas the top three spots
are occupied by Western states,Bulgaria, Slovakia, Slovenia, Poland, Latvia,
and the Czech Republic stresscitizenship more than do Spain and the Nether7. There are two options available for reporting the regional scores. One is to average the
country scores for each region. This ensures that each country receives the same weight in the
composite score. However, it is impossible to conduct any tests of statistical significance for
regional differences using such scores. The other option is to pool all the respondents for each
region and average their scores. The disadvantage here is that some countries are
overrepresented due to their larger sample sizes, whereas others are underrepresented. This article chooses the latter method so that the statistical significance of differences between the
regions may be assessed.Fortunately. the effects of overrepresentation and underrepresentation
are very small, because the regional scores for each of the two tnethods are extremely close for all
measurements in this article.

POLITICALSTUDIES/June2002
S66 COMPARATIVE
lands.8As well, Eastern Germanys score is lower than Western Germanys,
contrary to expectations.
Another question askedrespondents how important it is to feel that one is
a member of the nation. This is a good measureof the civic notion of will and
consent. Renans (1994) famous quotation about a nation being an everyday
plebiscite refers precisely to this voluntary, subjective nature of civic nationhood (p. 17). For people who adhere to a cultural or ethnic conception of the
nation, stateof mind is less important. More central are objective factors such
as descent and knowledge of a particular language. Whether an individual
strongly feels himself or herself to be an X or not, having parents of X ethnicity or speaking X makes him an X in the ethnic and cultural conceptions.
Table 4 indicates again that the dominant ethnic group in each country overall
finds will/consent an important element of national identity. Greatly contradicting the idea of a West much more civic than Eastern Europe is the dominance of Eastern European countries in the top live spots. Indeed, the Eastern
European average of 1.31 is lower than the Wests average of 1.58. Whereas
the higher scores in Central Europe compared with the West conform to the
conventional wisdom, the relatively high scoresof such Western countries as
Great Britain, the Netherlands, and Spain do not.
Given the newness of the political and legal systemsin all the statesof the
region, it would be surprising to find that Eastern Europeans deem respect for
political institutions and law an important criterion for national membership.
Yet as the data show in Table 5, the average score for Eastern Europe is just
1.77, between very imporrunrandfuirly important (although closer to the latter, of course). This score is higher than the Wests (1.48), but considering
that the age, stability, and effectiveness of Western political and legal institutions far outstrip those of Eastern Europe, the difference between these averagesis quite small. It is quite difficult for the standard civic-West/ethnic-East
paradigm to explain the fact that respondents in Latvias 4-year-old state (in
1995), plagued with economic problems and corruption, deem respect for
political institutions and law as important for nationhood as do respondents
in the 219-year-old United States, arguably the most stable and wealthy
country in the world (the .07 point difference here is not statistically significant). The same is true of Bulgarias .05 point difference with Great Britain,
8. Because there are more than 130 dyadic combinations of countries for each of the tables in
this article, reporting tests of statistical significance for the difference in average country scores
for each combination is not feasible. However, examination of these dyads using t tests reveals a
minimum level of difference between scores necessary to achieve statistical significance. For the
4-point scales, nearly all differences between countries that equal or exceed .I0 are significant at
the .05 level or better. For the Spoint scales, all differences in country averages that equal or
exceed .12 are significant.

567

Shuhnan I CJVJC/JZTHNJC AND WEST/EAST DICHOTOMIES


Table 4
Ethnic Majority Views on Importance for National
ber of the Nariotr
country

Hungary
Bulgaria
Poland
Slovakia
Czech Republic
Canada
Latvia
Slovenia
Norway
United States
Sweden
Spain
Great Britain
The Netherlands
West Germany
East Germany
West
Central Europe
Eastern Europe
Note: 1 = vety

importarrf;

4 = uot important

Membership

of

Feeling Oneselfto

be a Mem-

Mean

913
895
1519
1204
865
1056
552
929
1411
825
1311
971
818
1969
1126
561
8360
1687
6936

1.17
1.23
1.32
1.32
1.34
1.35
1.39
1.46
1.48
1.51
1.58
1.63
1.73
1.73
1.85
1.86
1.58
1.85
1.31

at all.

which also is not statistically significant. Furthermore, five Eastern European


countries are ranked higher than Spain and the Netherlands, and the differences are statistically significant in each case except the comparisons with
the Czech Republic. Thus although the regional averages for this item conform to the conventional view (civicness of West > Central Europe > Eastern
Europe), the absolute values of the Eastern European scores and their position relative to some Western states are inconsistent with that view.
Finally, the survey data can measuresupport for a territorial conception of
nationhood in the 16 cases.But first we must discuss how exactly territory is a
basis for national unity in a civic state.In his description of civic nation building in the Third World, Smith (1991) explains that living together and
being rooted in a particular terrain and soil become the criteria for citizenship and the basesof political community (p. 117). Presumably those who
have lived a long time in a state are more rooted in the territory and thus are
more authentic members than new residents. Similarly, someoneborn in a
territory may be seenasmore attachedto it than an immigrant. Using territory
as a criterion of national membership is inclusive in that anybody who is born

568

COMPARATIVE POLITICAL STUDIES /June 2002

Table 5
Ethnic Majority
Membership

Mews on Importance

of

Respect for Political

country

Sweden
Norway
Canada
United States
Latvia
West Germany
Great Britain
Bulgaria
East Germany
Slovakia
Slovenia
Czech Republic
Spain
The Netherlands
Poland
Hungary
West
Central Europe
Eastern Europe
Note: 1 = very important;

4 = uot important

Institutions

and Law for National

Mean

1351
1429
1058
823
552
1129
814
864
555
1187
921
832
965
1940
1451
934
8379
1684
6740

1.17
1.23
1.36
1.42
1.49
1.53
1.56
1.61
1.66
1.66
1.67
1.75
1.77
1.80
1.85
2.19
1.48
1.58
1.77

at all.

or lives a long time in a country meetsthis criterion-regardless of ethnicity,


religion, class, race, and so forth. It is exclusive, however, in that some (those
who are newly resident) or all immigrants are not considered fully part of the
national community.
The survey data indicate that dominant ethnic groups in the Eastern European states are generally more supportive of a territorial notion of national
identity than in the Western states,although both regions are supportive in an
absolute sense (see Tables 6 & 7). In fact, only one country, Canada, has a
majority of respondents who do not stressthe importance of one of the two
types of territoriality. In addition, both regions place slightly more importance on length of residence than on birth, the latter of course being the harder
standard to meet (i.e., it is more exclusive). Particularly interesting is the
strong emphasis given to territoriality by those majority ethnic groups that
have diasporas in other countries. For example, these data suggest that most
Germans in the two Germanies, Czechs in the Czech Republic, Hungarians in
Hungary, Slovaks in Slovakia, and so on do not think that their ethnic brethren in other countries in the region are real members of the nation-a sign

Shulman / CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

569

Table 6
Ethnic Majority
Membership

Views on Importance

of

Spending Most of Ones Life in Country for National

country

Bulgaria
Czech Republic
Hungary
Latvia
Poland
Great Britain
Spain
Slovenia
Slovakia
United States
Norway
East Germany
Sweden
West Germany
The Netherlands
Canada
West
Central Europe
Eastern Europe
Note: 1 = ve,y importartt;

4 = not important

Mean

882
862
963
554
1494
806
972
937
1204
825
1418
562
1322
1129
1971
1069
8384
1691
6897

1.65
1.73
1.79
1.81
1.82
1.84
1.84
1.86
1.89
1.93
1.99
2.04
2.15
2.17
2.27
2.34
2.09
2.13
1.80

at all.

that ethnic nationhood is weak in the region. Also, for those Westerners who
consider a stress on territoriality a mostly exclusive and negative phenomenon, the importance that Britain, Spain, and even the United States-the land
par excellence of immigration-place on birth must be disconcerting.
Just asthe survey questions show that civic sourcesof national identity are
stronger in Eastern Europe than is usually supposed, they also show that the
cultural bases of national identity are stronger in the West than is usually
supposed.
When asked about the role of speaking the countrys dominant language
in constituting nationhood, respondents in all countries, including those in
the West, generally think language is important, with country scoresranging
from 1.25 to 1.88 (seeTable 8). Although Eastern European scores (1.39) are
lower on average than Western ones (1.49), the difference is very small.
There are some unexpected scores from the standpoint of the civic-West/
ethnic-East stereotype. The United States, the Netherlands, Sweden, and
Norway all have scoresof 1.38 or lower. Also, five of the sevenWestern states
place greater importance on language than both halves of Germany, with all

570

COMPARATIVE POLITICAL STUDIES I June 2002

Table 7
Ethnic Majority

Views on Importance

of Being Born in Countryfor

National

Membership

country

Meal1

Bulgaria

892

1.54
1.71
1.79
1.87
1.94
1.95
1.96
2.01
2.09
2.11
2.14
2.22
2.35
2.38
2.39
2.54
2.20
2.32
1.90

830
1504
976
852
940
559
960
823
1205
1433
579
1994
1145
1346
1064
8466
1724
6911

Great Britain
Poland
Spain
Czech Republic
Slovenia
Latvia
Huogary
United States
Slovakia
Norway
East Germany
The Netherlands
West Germany
Sweden
Canada
West
Central Europe
Eastern Europe
Note: 1 = vety important;

4=

notimportant

nt all.

differences statistically significant. Becauselanguage has historically played


such an important role in German identity, this is particularly good evidence
of the depth of culturally grounded national identities in the West.
The ISSP survey also asked about the role of religion in national identity
(see Table 9). In contrast to all the previous items, in most of the 16 cases,a
majority of respondents does not consider religion to be an important criterion for national membership (scoresexceed 2.5). The data on the importance
of religion present a similar ranking pattern as does language, with Eastern
Europe leading the pack, followed by the West, and then Central Europe.
However, several Western states again scored lower than several Eastern
European states.TheUnited States,Spain, and Britain all place greater stress
on religion than Slovenia, Slovakia, and the Czech Republic. The nearly
identical scores of the supposedly civic United States and ethnic Poland are
particularly noteworthy, especially given the key role Catholicism and the
Catholic Church have played in the national revival in Poland during and
since communism.
A final measureof the cultural component of national identity comes from
the following: Now we would like to ask you a few questions about minori-

Shuhnan / CIVIC/ETHNIC AND WEST/EAST DICHOTOMIES

571

Table 8
Ethnic Majority
Membership

Views on Importancr

of Ability

to Speak Dominant

Country

Hungary
Czech Republic
Norway
Slovakia
Sloveuia
Sweden
Latvia
The Netherlands
United States
Bulgaria
Great Britain
Poland
West Germany
East Germany
Spain
Canada
West
Central Europe
Eastern Europe
Note: 1 = very importarlt;

4 = aot importmt

Language for National

Meau

970
869
1434
1209
946
1353
555
2005
832
889
828
1519
1144
576
972
1068
8492
1720
6957

1.25
1.29
1.30
1.33
1.33
1.33
1.36
1.38
1.38
1.41
1.47
1.57
1.57
1.62
1.87
1.88
1.49
1.59
1.39

at ~11.

ties in [respondents country]. How much do you agree or disagree with the
following [statement]: It is impossible for people who do not share the customs and traditions [of respondents country] to become fully [e.g., British,
German, Hungarian, etc.]. Respondents used a 5-point agree/disagree scale
(agree strongly, agree,neither agreenor disagree,disagree,disagree strongly).
Note that in contrast to the previous questions, a score less than 3.0, not 2.5,
indicates that a majority of respondents finds this element of national identity
important (i.e., agrees with the statement). In only 3 of the 16 cases does a
majority not agreethat sharing the traditions and customs of acountsy is necessary for national membership (see Table 10). Consistent with the civicWest/ethnic-East argument. Eastern Europe (mean = 2.39), on the whole,
emphasizes customs and traditions more than the West (2.62), although Norway, the Netherlands, and Sweden place more importance on this item than
four Eastern European countries, with all differences statistically significant.
Again, the difference between the West and Eastern Europe, although statistically significant, is small: .23 point on a 5-point scale. In addition, the two

572

COMPARATIVE POLITICAL STUDIES /June 2002

Table 9
Ethnic Majority
Membership

Views on Importance

ofBeing

a Believer

country

Bulgaria
United States
Poland
Spain
Hungary
Latvia
Slovenia
Great Britain
West Germany
Slovakia
Canada
Norway
Czech Republic
East Germany
Sweden
The Netherlands
West
Central Europe
Eastern Europe
Note: 1 = very importarrt;

4 = rrol important

in Domimm~ Religion for National

Meao

862
816
1455
965
960
525
913
810
1112
1176
1036
1328
832
542
1297
1945
8198
1654
6724

1.80
2.41
2.43
2.64
2.79
2.80
2.82
2.83
2.92
3.09
3.12
3.13
3.14
3.20
3.30
3.58
3.10
3.01
2.69

at (111.

Germanies place slightly less importance on this cultural component (2.88)


than do Western countries+ontrary to established views.
This articles analysis of the ISSPs direct measures of national identity
permits several conclusions. First, for every measure,there are some Eastern
European countries that are more civic and less cultural than some Western
countries. Second, the general pattern of support in the West, Eastern Europe,
and Central Europe for the civic and cultural content of identity is similar.
Table 11 demonstrates that a majority of respondents in all three regions
agree on whether or not a given component of national identity is important.
For example, on the question on citizenship, the average scores of the Westem (1.63, Eastern European (1.69), and Central European (1.77) samples
are all less than 2.S-the midway point on the 4-point scale-signifying that
most people in each region find this component of national identity important. This pattern of basic regional agreement is found for all eight direct
measuresof national identity in the study. These results should caution scholars against characterizing these three regions as fundamentally different in
their concepts of nationhood.

Shulman I CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

513

Table 10
Ethnic Majority Views on Need for Minorities
Become Members of Nation

Country
Bulgaria
Latvia
Norway
Slovenia
The Netherlands
Sweden
Great Britain
Hungary
Czech Republic
Poland
East Germany
Spain
West Germany
Slovakia
United States
Canada
West
Central Europe
Eastern Europe
Note: 1 = agree stmngly;

to Share Country vS Customs and Tradiriorzs to

Mean

896
534
1414
908
1965
1307
821
938
830
1354
528
898
1080
1185
782
1063
8251
1608
6645

1.35
1.96
2.21
2.33
2.36
2.37
2.52
2.54
2.55
2.56
2.79
2.85
2.92
3.00
3.09
3.39
2.62
2.88
2.39

5 = disagree strongly.

The civic-West/ethnic-East argument can also be e+aluated by more narrowly comparing the intensity of support for a given component of national
identity in the three regions. Elements of this argument that are confirmed by
the direct survey measuresare marked by an X in Table 12. Comparing the
West and Eastern Europe (Column I), the standard argument correctly predicts greater support in Eastern Europe for the three cultural components of
national identity, but correctly predicts greater support in the West for civic
identity in just two of the five components--citizenship and respect for political institutions and law. Furthermore, for each component, the differences in
average scores for the two regions (shown in parentheses) are small. The
standard interpretation also predicts a stronger civic and weaker cultural
identity in the West than in Central Europe. The data support this claim in all
of the civic items (one of which, Territory-Long Residence, is not statistically significant) but in just one of the cultural items (Column II). We should
also find the two Germanies to have stronger civic and weaker cultural identities than Eastern Europe. This holds true for all the cultural items but for just
one of the civic items (Column III). Next, the two German& provide a nice

574

COMPARATIVE POLITICAL STUDIES /June 2002

Table 11
Regions Where Majorigv ofRespondents

Finds Given Component ofNational

Component of National Identity


Civic
Will
Citizenship
Respect insritutioosflaw
Territory-birth
Territory-long residence
Cultural
Language
Religion
Traditions

Identity Importatzt

West

Ceutral
Europe

Eastern
Europe

X
X
X
X
X

X
X
X
X
X

X
X
X
X
X

Note: X indicates that the regional average is less than 2.5 for each item (except Traditious,
where average is less than 3.0).

quasi-experiment on the effect of communism on national identity. The standard argument predicts stronger civic and weaker cultural identity in West
Germany than in East Germany. However, only three of the eight direct measures of national identity conform to this expectation, and one of those (Will)
is not statistically significant (Column IV). In all, of the 32 elements of Ihe
civic-West/ethnic-East stereotype evaluated in Table 12, only 16 are confirmed by statistically significant measures.And even for these 16 elements,
the regional differences are very small. The averagepoint differential in these
16 instances for the measuresusing a 4-point scale is .18, and the averagedifferential for the 5-point Tradition measure is .28.
INDIRECT

MEASURES OF NATIONAL

IDENTITY

In addition to investigating support for various contents of national identity by looking at massattitudes toward criteria for membership in the nation,
we can examine attitudes toward key policy issues: cultural assimilation and
immigration.
Two questions from the ISSP survey assessthe majoritys attitude toward
minority assimilation and the states role in promoting or resisting it. One
question read,
Some people say that it is better for a country if different racial and ethnic
groups maintain their distinct customs and traditions. Others say that it is better

Shulman / CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

57s

Table 12
Elenzents of the Civic- West/Ethnic-East Argument
National Identity (indicated by X)

Component of
National Identity
Civic
Will
Citizenship
Respect institutions/law
Territory--birth
Territory-long residence
Cultural
Language
Religion
Traditions

ThatAre

Confirmed by the Direct Measures of

1
W vs. EE

II
w vs. CE

111
CE vs. EE

IV
WG vs. EG

(.27)***
x (.04)***
x (.29)***
(.30)***
(.29)***

X
x
x
x
x

(.s4)***
(.08)***
x (.19)***
(.42)***
(.33)***

x (.Ol)
(.09)*
x (.13)***
(.16)**
(.13)**

x (.lo)***
x (.41)***
x (.23)***

(.lo)***
x (.09)**
(.26)***

x (.20)***
X (.32)***
x (.49)***

W)

(.27) ***
(.12)***
(.lo)***
(.12)***
C.04)

(.28,***
x (.13)*

Note: I: Civic identity W > EE, Cultural identity EE > W; II: Civic identity W > CE, Cultural

identity CE > W, III: Civic identity CE > EE, Cultural identity EE > CE; IV: Civic identity WC >
EG, Cultural identity EG > WG. W = West; EE = Eastern Europe; CE = Central Europe: WG =
West Germany; EG = East Germany. Point differentials between regional averages are given in
parentheses.Baaed on Cpoint scales for all items except Traditions, which used a 5-point scale.
*p<.o5.**p~.o1.***p<.oo1.

if thesegroupsadaptandblend into the larger society.Which of theseviews


comescloserto your own?
1. It is betterfor societyif groupsmaintaintheir distinct customsandtraditions.
2. It is betterif groupsadaptand blend into the larger society.
This question ascertains the respondents general support for assimilation.
The results show striking differences between the Western and Eastern European countries in the sample, with the former preferring minority assimilation at substantially higher rates than the latter (seeTable 13). The countries
ranked highest in support for assimilation are all Western countries, whereas
the countries ranked lowest are all Eastern and Central European countries.9
A majority of respondents in every Western country except Spain desires
assimilation of minorities into the dominant culture. Based on the earlier discussion, this result is good evidence that cultural national identity is strong in
the West.In addition, majorities in every Eastern and Central European country except Bulgaria are against assimilation. But one must be careful in draw9. Chi-square testing indicates that all differences between countries in support for assimilation equal to or greater than five percentage points are st&tically significant.

576

COMPARATIVE POLITICAL STUDIES / June 2002

Table 13
Ethnic Majorir);

country
Latvia
East Germany
Hungary
Slovakia
Slovenia
Poland
Spain
West Germany
Czech Republic
Bulgaria
United States
Canada
The Netherlands
Norway
Great Britain
Sweden
West
Central Europe
Eastern Europe

Attitudes

Toward Cultural Assimilation

N
483
451
894
1068
766
956
856
897
130
780
599
889
1660
1208
702
1150
7063
1348
5671

ofMinorities
% Against
Assimilation

70For
Assimilation

70
61
60
58
54
52
52
52
50
48
40
36
29
24
19
18
30
55
55

30
39
40
42
46
48
48
48
50
52
60
64
71
76
81
82
70
45
45

ing inferences from this with regard to national identity. Resistance to assimilation may be a sign of ethnic identity or civic identity. Eastern and Central
European respondents, who represent majority ethnic groups, might think
that minorities can never become part of the nation regardless of what culture
they adopt, becausethey are of the wrong ancestry or race. Thus resistance to
assimilation may reflect exclusionary attitudes based on ethnicity.
Another measure of mass policy preferences in the cultural sphere does
allow distinguishing between support for ethnic and civic concepts of nationhood. Respondents in each country were asked how much they agree or disagree with this statement (using a 5point agree-disagree scale): Ethnic
minorities should be given government assistanceto preserve their customs
and traditions. Yet again, the data are highly inhospitable to the standard
civic-West/ethnic-East argument (seeTable 14). The five most multicultural
countries are Eastcm and Central European, whereas the six least multicultural are Western. For all the talkof multiculturalism in the West, particularly
the United Statesand Canada, the weak desire to aid ethnic minority cultures
in these countries relative to post-communist Europe poses a great challenge
to any scholar adhering to the conventional wisdom on ethnic and civic

Shulman /CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

577

Table 14
Ethnic Majority

Supportfir

Govenunent

Assistance

countrv

Hungary
Slovenia
Poland
East Germany
Latvia
Spain
Slovakia
Czech Republic
Bulgaria
West Germany
The Netherlands
Sweden
Norway
Great Britain
United States
Canada
west
Central Europe
Eastern Europe

to Preserve Minority

Cultums

Mean

954
819
1395
552
520
877
1152
800
828
1070
1925
1245
1374
812
784
1050
8067
1622
6527

1.91
2.26
2.28
2.34
2.44
2.54
2.65
2.86
2.99
2.99
3.41
3.46
3.56
3.69
3.15
3.84
3.47
2.77
2.46

Note: 1 = agree strongly; 5 = disagree strongly.

nations. A desire for governmental support for ethnic minority cultures is not
consistent with an ethnic national identity centered on the dominant or titular
group.
Another major policy issue that can shed light on conceptions of national
identity in a country is immigration. ISSP respondents were asked the
following:
Do you think thenumberof immigrantsto [respondentscountry] should. .
1. Be increaseda lot.
2. Be increaseda little.
3.
4.
5.

Remain the same as it is.


Be reduced a little.
Be reduced a lot.

Civic identities should lead to greater opennessto immigration, whereas cultural and ethnic identities should be more restrictive. Unfortunately, the data
do not permit taking into account how the cultural or ethnic background of

578

COMPARATIVE POLITICAL STUDIES I June 2002

Table 15
Ethnic Majority

Supportfor

Increusing/Decreasing

country

Spain
Canada
The Netherlands
Poland
Norway
United States
Slovenia
Sweden
Slovakia
Great Britain
Czech Republic
Bulgaria
West Germany
East Germany
Hwruy
Latvia
West
Central Europe
Eastern Europe

Immigratiorz

Rates
N

Mean

856
950
1888
1092
1363
723
893
1272
1001
800
789
608
1063
541
935
511
7852
1604
5828

3.40
3.43
3.84
3.84
3.86
3.92
3.99
4.01
4.02
4.09
4.17
4.23
4.26
4.34
4.42
4.46
3.81
4.28
4.13

Note: 1 = increased a lot; 5 = decreased a lot.

immigrants affects respondents attitudes toward their entry. As the data demonstrate (seeTable 15), most respondentsin all countries prefer that immigration be scaled back. To the extent that attitudes toward immigration reflect
national identity, the data for the most part conform to the standard civicWest/ethnic-East interpretation. The averagescore of Eastern Europe is 4.13,
whereas that of the West is 3.81. Although the difference is in the expected
direction and statistically sign&ant, the .32 gap is small considering the
5-point scale used. Not consistent with the standard view is the fact that the
two Germanies express more resistance to immigration than does Eastern
Europe. Given that most Westernerswish to scale back immigration, the best
interpretation of this measurement of identity is not that the West more
strongly supports civic nationhood than Eastern Europe, but that it is slightly
less supportive of ethnic or cultural nationhood.
The indirect measures thus point in different directions from the standpoint of evaluating the civic-West/ethnic-East argument. Consistent with the
conventional wisdom, attitudes toward immigration point to the greater
strength of cultural or ethnic identities in Eastern Europe than in the West. Yet
attitudes toward assimilation and multiculturalism contradict the standard

Shulman I CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

519

argument, pointing toward stronger civic and weaker cultural identities in


Eastern Europe than in the West. In terms of the normative concerns of many
scholars who employ the standard civic/ethnic and West/East dichotomies,
Eastern Europe is more tolerant of its internal minorities than is the West,
whereas the West is more tolerant of immigrant minorities than is Eastern
Europe. In addition, East Germany demonstratesgreater support for internal
minority cultures than does West Germany, yet another indication that experience with communism does not seemto stimulate the cultivation of ethnic
and cultural national identities. Although East Germany is less welcoming of
immigrants than West Germany, the difference is extremely small (.08) and
not statistically significant. Like the direct measuresof national identity, the
indirect measuresoverall present a much more complex picture of national
identification in the three regions than portrayed by the civic-West/ethnicEast argument, with the West more cultural than is usually supposed and
Central and Eastern Europe more civic.

EXPLAINING STRONG CULTURAL


NATIONAL IDENTITY IN THE WEST
Why should it come as no surprise that majority ethnic groups in the Westem countries harbor a relatively strong cultural national identity, especially
as reflected in their views toward language and cultural assimilation? There
are several reasons.
First, both of the main sides in the debateover the origins and modernity of
nations imply a strong role for culture as a basis for national identity in the
West. Smiths (1986) argument that nations have ancient ethnic cores
around which nations are constructed clearly demands a great role for the cultural components of national identity. A nation founded on the culture, symbols, myths, and memories of a dominant ethnic group is likely to retain its
culturally based identity for a very long time. Smith writes that the earliest
casesof the Western model (England, France, Spain, Holland, Sweden, Russia) were ethnic states that were gradually transformed into nations through
economic unification, territorial centralization, provision of equal legal
rights, and growth of mass public education (p. 138). It is interesting that
Kohns argument begins with assumptions similar to Smiths. Recall that in
Kohns Western model, the state encompasseda relatively culturally homogeneous population, which the state then forged into a nation. The strong
match between ethnography and political boundaries in Kohns Western
model is akin to Smiths ethnic core of modern nations. But Kohn did not follow his initial assumption to its logical conclusion-that relative cultural

5230

COMPARATIVE POLITICAL STUDIES / June 2002

homogeneity in a state would be a strong force binding the people together


into a nation. By the logic of both Kohns and Smiths approaches, since the
Western nations inception, culture has been prominent in uniting and distinguishing a large proportion of the population of a state.
Gellner (1983) and other modernists disagree with Smiths contention
about the ethnic origins of nations. Gellner instead explains the rise of nations
through a functional argument whereby states intentionally forged national
unity to meet the demands of industrial development. The state, primarily
through public education, pursued cultural homogenization of its population
to create a mobile workforce able to communicate with strangers. Another
modernist, Benedict Anderson, focuses not on industrialization but the rise of
print capitalism in forging national consciousness. In Andersons (1983)
view, mass communication stimulated the diffusion of a common culture
based on a vernacular language. In each theory, culture plays a key role in
uniting a large proportion of a states population, just as in Smiths version.
The difference is primarily one of timing. If Smiths argument means that
cultural and civic elements should coexist from the start of Western states
national development, Gellners and Andersons arguments imply that the
cultural elements should be weak at the start of national development and
then gain strength as industrialization and mass communications developed
over the 19th and 20th centuries.
A second explanation for the current strength of the cultural components
of national identity in the West centers on the inability of civic components of
national identity to provide sufficient unity in a state.To the extent that a civic
identity is rationalistic and voluntary, it is unlikely to cultivate an attachment
to the nation with the great emotional resonance that ethnic and cultural elements provide. But the main reason that civic components of nationhood fall
short in their ability to evoke emotional attachment to the nation is because
most civic components of nationhood are external to the individual, whereas
ethnic and cultural components are internal. Territory, political institutions
and rights, and citizenship exist outside the individual, whereas ancestry,
race, religion, language, and traditions are a direct part of a persons physical
and psychological makeup. As a result, the intensity of attachment to communities founded predominantly on the latter will likely exceed those
founded predominantly on the former. The dismal record of civic-territorial
nation building in many postcolonial multiethnic statesin the Third World is
a testament to the weaknessof civic nationalism unsupported by elements of
ethnic or cultural nationhood at the statewide level. Using functionalist logic
similar to Gellners, a case can be made that Western states for most of the
past two centuries have promoted a homogeneous linguistic and cultural
identity precisely due to the ability of culture to provide cohesion for popula-

Shulman I CIVIC/ETHNIC

AND WEST/EAST DICHOTOMIES

581

tions in an environment in which civic elements of nationhood alone were not


up to the task.
Finally, all social identities, including national ones, require that the social
unit be differentiated from other units. Purely or predominantly civic-based
national identities are unable to provide a sufficiently high degreeof differentiation in the modem world. The very universalistic nature of civic nationalism that many applaud is a reason for its inability to provide the sole or even
overwhelming basis for identification in the West. Not only is it relatively
easy for states to make citizenship an important criterion for national membership, but it is exceedingly common. Similarly, all stateshave territory, and
making attachment to that territory a criterion for national membership is
also easy and ubiquitous. There is more room for national variation in political rights and institutions, and belief in political principles. However, the
spread of democratic government and ideology in the past two centuries has
greatly reduced this variation. In this context, culture, which varies substantially from place to place, grows in importance for its ability to distinguish
one Western nation from another and from non-western nations.

EXPLAINING CIVIC NATIONAL


IDENTITY IN EASTERN EUROPE
How is it that the data show such strong support for civic components of
national identity among majority ethnic groups in Eastern Europe?
Those observers who expect nationalism in Eastern Europe to be mostly
of the ethnic variety neglect to take into consideration the enormous influence and prestige of the civic model of the nation. Whereas Western nations
have long been basedon culture, the rhetoric emanating from these countries
stressing the centrality of territory, citizenship, and political institutions to
nationhood has been strong. This civic conception of the nation diffused to
the non-West, in large part due to wealth and power of the West. Western
ideas at the end of the 18th century about the state being the representative
and embodiment of the nation have become commonplace, precisely because
of the military and economic success of what is presented as the civicnation-state model. In this context, ethnonational groups look to statehood,
or at least a degree of political autonomy within an existing state, as a means
of defending and promoting their culture, identity, and interests. Thus as soon
as Hungarians, Czechs, Latvians, and so forth become majority and titular
groups in independent states, free of Soviet domination, attachment to and
membership in the nation quickly become measured against a persons relation to the state-its territory, institutions, laws, and the like.

582

COMPARATIVE POLITICAL STUDIES / June 2002

A second, and related, reason for the strength of civic-based national identities in Eastern Europe is the historical experience of many of thesecountries
with statehood. Of the seven Eastern European titular groups in this sample.
five have enjoyed independent statehood prior to indirect or direct Soviet
domination: Bulgarians, Poles, Hungarians, Czechs, and Latvians. The other
two groups, Slovenians and Slovaks,attainednominal sovereignty in territorialadministrative units of the Yugoslavian and Czechoslovakian federations,
respectively. Consequently, contemporary national identity did not develop
solely in the context of struggle against political borders in Eastern Europethe dimension stressedby Kohn and others. It also developed around political
borders that overlapped to a large degree with ethnic and cultural borders.
Furthermore, Eastern European ethnic groups national mythologies celebrate their earlier periods of statehood as golden ages. With the collapse of
communism, the independent state and its institutions and territory are
looked to as vehicles of national renaissance.
Finally, another aspect of Kohns original argument provides an insight
into the strength of civic national identity in Eastern Europe. Kohn wrote that
nationalism in Central and Eastern Europe developed in reaction to the
French invasions during the Napoleonic Wars and the universalist ideology
of Western liberalism. To counter this, Central and Eastern Europe looked to
their own unique historical and cultural heritage as the basis of national identity. This same argument can be adapted to post-Communist Eastern Europe
to explain the strength of civic components of national identity there. Modem
national identities in this region developed in opposition to Soviet domination and its universalist communist ideology. As Soviet and communist rule
weakened and finally collapsed, people in the region again rallied around
their heritage, but now they interpreted this heritage as fundamentally
democratic-in contrast to that of their imperial overlord. In constructing
national identity and asserting national autonomy, most titular ethnic groups
in the region consequently underscored democratic principles flouted during
the communist period-the rule of law, political equality, and minority
rights. Therefore, whereas the reactive nature of Eastern European nationalism in the 19th century may have led to a stresson ethnic and cultural components of national identity, in the late 20th century it led to a stress on civic
components.

CONCLUSION
The empirical analysis in this article demonstrates that the traditional
civic-West/ethnic-East argument is a gross simplification of concepts of

Shulman / CIVICJETHNIC AND WEST/EAST DICHCrrOMIES

583

nationhood in the West, Central Europe, and Eastern Europe. Although some
survey measuresexamined herein support the argument, an equal number do
not. Furthermore, even when the argument is true, it is only weakly true, as
the differences in national identity between the regions are small and there is
a substantial degree of diversity within each region. Overall, the data suggest
that imperial and communist rule have not pushed Eastern European nationhood in a strongly cultural direction while greatly weakening civicness. And
whereas most of the West has a long tradition of democracy and relatively
strong and stable political institutions, cultural conceptions of nationhood are
alive and well, and support for multiculturalism is relatively weak. The data
reveal an interesting tension between policies adopted by many of the states
in the sample and the identities of their inhabitants. For example, although
Canada has adopted an official policy of multiculturalism, Anglophones
there are more supportive of minority assimilation than are majority groups
throughout Central and Eastern Europe. Likewise, although Germany until
1999 had a citizenship law more ethnically discriminatory than that of any
Westerndemocracy, ethnic Germans themselves place greater importance on
developing minority cultures and less importance on language and traditions
as criteria of national membership than do most other Western majority
groups in this study. This suggeststhat scholars must be careful not to overestimate the ability of states to shape mass national identities.
Last, this article has several limitations that must be noted. First, it examines current patterns of national identity. It is possible, even likely, that historically the groups under study here exhibited different degrees of ethnic/
cultural/civic components in their national identity than they do now. Second,
the analysis relies on just one dataset from 1995-1996 that may reflect
short-term trends away from long-term patterns more consistent with the
conventional wisdom. For example, ethnic conflict in Yugoslavia and the former USSR may have temporarily delegitimized ethnic and cultural notions of
national identity in Central and Eastern Europe. Third? patterns of support for
ethnic/cultural versus civic notions of national identity may systematically
differ for masses and elites, with elite views more consistent with the
civic-West/ethnic-East pattern. Further study of a wide range of political
elites in each state will tell if there is a mass-elite gap in identification on this
score. Fourth, the limitations of the survey data did not permit a rigorous
assessmentof the differences in support for the strictly ethnic content of
national identity in the casesunder study, or analysis of the role of political
principles in generating civic identities. Additional multicountry surveys
should be conducted to ascertain whether ancestry, race, and political ideology play a different role in national identity in the West, Central Europe, and
Eastern Europe. Finally, much work remains to be done in explaining the dif-

5x4

COMPARATIVE POLITICAL STUDIES / June 2002

ferences in national identity found among the individual countries in this


study. Factors such as the settlement patterns of a nations main ethnic
groups, the cultural distance between these groups, the existence and size of
diaspora groups outside the country, and levels of immigration all may affect
the strength of ethnic/cultural/civic conceptions of national identity and thus
merit study.

REFERENCES
Alter. Peter. (1994). Nationalism (2nd ed.). London: Edward Arnold.
Anderson. Benedict. (1983). Imagined communities: ReJections on the origins and spread of
nationalism. London: Verso.
Brubaker, Rogers. (1992). Citizenship and nationhood in France und Germany. Cambridge,
MA: Harvard University Press.
Brubaker, Rogers. (1996). Nationalism reframed:
Nationhood and the national question in
Europr. Cambridge: Cambridge University Press.
Calhoun. Craig. (1997). Narionalism. Minneapolis: University of Minnesota Press.
Gellner, Ernest. (1983). Nations and natiomdism. Ithaca, NY: Cornell University Press.
Gellner, Ernest. (1992). Nationalism in a vacuum. In A. J. Motyl (Ed.), Thinking theoretically
about Soviet nntiorurlitius: Hisrory and comparison in the study of the USSR (pp. 243-254).
New York: Columbia University Press.
Ignatieff, Michael. (1993). Bloodand belonging: Journeys info the Ned nationalism. New York:
Farrar. Straus, Giroux.
Kohn, Hans. (1944). TIte idea ofnationalism: A study in its origins and background. New York:
Macmillan.
Kohn, Hans. (1946). Prophets and peoples: Studies in nineteenth century nutionnlism. New
York: Macmillan.
Kohn, Hans. (1949). The tnxentieth century: A mid-waynccortnt ofthe western world. New York:
Macmillan.
Kymlicka, Will. (1999). Misunderstanding nationalism. In R. Beiner (Ed.), Theorizingnutionalism (pp. 131.140). Albany: State University of New York Press.
McCrone, David. (1998). The sociology ofnationalism.
London: Routledge.
Meincke.Friedrich. (1970). Cosmopolifunism~dthenationnlstute
(7th ed.,RobertB. Kimber,
Trans.). Princeton, NJ: Princeton University Press.
Nieguth, Tim. (1999). Beyond dichotomy: Concepts of the nation and the distribution of membership. Nations and Nutiorudism, 5, 155-173.
Nielsen, Kai. (1999). Cultural nationalism. neither ethnic nor civic. In R. Beiner (Ed.), Theorizing nutionalism (pp. 119-l 30). Albany: State University of New York Press.
Renan, Ernest. (1994). Quest-ce quune nation? [What is a nation?]. In J. Hutchinson & A. D.
Smith (Eds.), Nationalism (pp. 17-18). Oxford: Oxford University Press.
Schopflin, George. (1995). Nationalism and ethnicity in Europe, east and west. In C. A. Kupchan
(Ed.), Nationalism andnationulities
in the RPWEurope (pp. 37-65). Ithaca, NY: Cornell University Press.
Schopflin, George. (1996). Nationalism and ethnic minorities in post-communist Europe. In
R. Caplan & J. Feffer (Eds.), Europes new nationaLism (pp.150-170). New York: Oxford
University Press.

Shulman / CIVIC/ETHNIC AND WEST/EAST DICHOTOMIES

585

Smith, Anthony D. (1986). 77~ ethnic origins ofnations. Oxford: Blackwell.


Smith. Anthony D. (199 I j. Nntional ictenfity. Reno: University of Nevada Press.
Snyder, Jack. (1993). Nationalism and the crisis of the post-Soviet state. In M. E. Brown (Ed.),
Ethnic conflicrund infemationrrlsecurity
(pp. 79-101). Princeton. NJ: Princeton University
Press.
Yack. Bernard. (1999). The myth of the civic nation. In R. Beiner (Ed.), Theorizing nationalism
(pp. 103-I 18). Albany: State University of New York Press.

Stephen Shulman is an assistant professor in the Department of Political Science at


Southern Illinois Universify, His wsearch interestsfocus on wtionalism
andethnicpolitics. He has an area specialtatiotz
in the former Soviet Union, where he has conducted
reserrrch and taught. His research appears in journals such as Europe-Asia Studies,
Political Geography, Ethnic and Racial Studies, nnd International Studies Quarterly.